Граф коммитов

1167 Коммитов

Автор SHA1 Сообщение Дата
Jeff Walden bae8a13c6e Bug 1341951 - Remove mozilla::AlignedStorage, and inline its sole use into js::jit::RInstructionStorage. r=nbp 2017-02-22 18:27:51 -08:00
Jeff Walden fceda02d9b Bug 1287006 - Use |alignas(T) unsigned char mStorage[sizeof(T)]| instead of AlignedStorage2 inside Maybe. r=froydnj
--HG--
extra : rebase_source : 59e8039481b332decacd1acc7cf1cb73ec8b78cf
2017-01-30 15:56:04 -08:00
Emanuel Hoogeveen 8867f10c5a Bug 1338574 - Assertions.cpp can't use MOZ_CRASH_ANNOTATE, so set the crash reason directly. r=froydnj 2017-02-22 15:42:00 -05:00
Jeff Walden d73ce21741 Bug 1287006 - Adjust media/ code to not pass Maybe (or any class containing a Maybe member) by value, only by reference or pointer. r=jw_wang, r=rjesup
--HG--
extra : rebase_source : 176a2afde5772bbdf43f0f6dc8082201acadcf4f
2017-02-13 09:07:40 -08:00
Jeff Walden 8529eb5317 Bug 1338374 - Shrink Vector from (usually) four pointers in size to three when no inline storage is used. r=froydnj
--HG--
extra : rebase_source : bc2c7f706305d04954162720b011029c8095229f
2017-01-30 15:56:05 -08:00
Jeff Walden d5deea2a4f Bug 1338374 - Make Vector not use AlignedStorage for its inline element storage. r=froydnj, r=keeler
--HG--
extra : rebase_source : 81eb5278404b4843ed5a59819f6000f74200aa41
2017-01-30 15:56:05 -08:00
Jeff Walden 8f8272ae04 Bug 1339555 - Make various operator-news into a known-non-null pointer use a ::operator new overload that odesn't null-check. r=froydnj
--HG--
extra : rebase_source : cad3ffa778db607bb2ccee2ac42671e60d3c8160
2017-02-14 11:23:18 -08:00
Emanuel Hoogeveen 9c7c14c872 Bug 1338574 - Part 1: Add MOZ_CRASH_UNSAFE_OOL and MOZ_CRASH_UNSAFE_PRINTF to crash with a runtime generated explanation string. r=froydnj
--HG--
extra : rebase_source : 3ba633aa77ab96f763ea03a8308c539c9c2aeda0
2017-02-21 18:01:41 +01:00
Emanuel Hoogeveen 4308a12546 Bug 1338574 - Part 0: Pass __LINE__ as an argument to MOZ_REALLY_CRASH() instead of using it directly. r=froydnj
--HG--
extra : rebase_source : 1df7776a4627c4af417be78649ba22203dca8026
2017-02-21 18:01:38 +01:00
Tom Tromey b3223b115f Bug 1060419 - fix comment in IntegerPrintfMacros.h to reflect Printf.h change, r=froydnj
MozReview-Commit-ID: 1Zdi7XNHUHL

--HG--
extra : rebase_source : f9c3cfd96504b232ad25cf1c49f3fcb07f461cc1
2016-12-14 14:16:20 -07:00
Jeff Walden d7372e37d9 Bug 1338374 - Use alignas/alignof to define Variant's internal raw storage. r=froydnj
--HG--
extra : rebase_source : a75613ba7c17ceccf20e6f096a118bfedfa1eb15
2017-01-30 15:56:05 -08:00
Jeff Walden 1811cee01b Bug 1338374 - Make tl::Min/Max variadic. r=froydnj
--HG--
extra : rebase_source : 140441135e9ee1fcde0541e39a19e1a5c47203c8
2017-01-30 15:56:05 -08:00
Jeff Walden 318b206490 Bug 1338374 - Use alignas/alignof, and don't use sizeof where alignof should have been used, in MaybeOneOf. r=froydnj
--HG--
extra : rebase_source : eb102c6344a6b2fc5a8c904d3361ac3f6735dd43
2017-01-30 15:56:05 -08:00
Masatoshi Kimura a35278e9a7 Bug 1325368 - Assume Win7 or later when testing Windows version. r=jimm
MozReview-Commit-ID: 718G0htR17H

--HG--
extra : rebase_source : e6f7a3584f03c802218a1de526345e1df1e67cf8
2017-01-29 16:42:46 +09:00
Nicolas B. Pierron 06bb70ebad Bug 1318677 part 3 - mozilla::Result: Add a new packing strategy to pack small enumerated values in a single word. r=Waldo 2017-02-07 18:57:43 +00:00
Nicolas B. Pierron 60cb2de0a8 Bug 1318677 part 2 - mozilla::Result: Use a single enum to dispatch to the Result implementation. r=Waldo 2017-02-07 18:57:43 +00:00
Nicolas B. Pierron cc14318018 Bug 1318677 part 1 - Use a smaller mask to remove the error tag. r=jandem 2017-02-07 18:57:42 +00:00
Jonathan Watt 33a6ca0109 Bug 1335780, part 2 - Tests for Maybe<const Type>. r=froydnj 2017-01-03 08:53:05 +00:00
Jonathan Watt 94c9b96404 Bug 1335780, part 1 - Make Maybe<T>::emplace() work when T is const. r=froydnj 2017-01-03 08:53:05 +00:00
Jeff Walden 3c1ae9aa91 Bug 1332797 - Delete the temporary directory used to store the double-conversion clone when all's said and done. r=froydnj
--HG--
extra : rebase_source : 838b42e130e757aae6795b41f75e0d04f0644a65
2017-01-26 13:26:40 -08:00
Jeff Walden cd89862c7a Bug 1332797 - Note the double-conversion git revision last used when importing double-conversion code. r=froydnj
--HG--
extra : rebase_source : 16af43fcda79ffc4743fda12c9ce945ee4fd8a26
2017-01-24 12:56:06 -08:00
Jeff Walden 5a966975cc Bug 1332797 - Update our double-conversion import using the command |./update.sh d8d4e668ee1e6e10b728f0671a89b07d7c4d45be|. r=generating-script-was-reviewed
--HG--
extra : rebase_source : 484d0ba0085c70b1e75790a098cbe4a1a9a0c95b
2017-01-24 09:30:13 -08:00
Jeff Walden 0fdc1cbc93 Bug 1332797 - Make update.sh fail if applying a local patch creates a *.{orig,rej} file. r=froydnj
--HG--
extra : rebase_source : 0b774ad1df5a8bc95d4af806f80bb7a2f6f8d738
2017-01-24 09:29:50 -08:00
Jeff Walden dd137bcf96 Bug 1332797 - Update our various local patches to apply against tip double-conversion. r=froydnj
--HG--
extra : rebase_source : 6b785879615aea4b9be0c31edb51bc86f8c958f9
2017-01-24 09:29:47 -08:00
Jeff Walden bdf8fd6644 Bug 1332797 - Adjust update.sh to work against tip double-conversion. r=froydnj
--HG--
extra : rebase_source : b34256579fe5e55fa1a3a7a2e4ae1495265055f4
2017-01-23 15:13:21 -08:00
Jeff Walden 645ca06a98 Bug 1332797 - Move mfbt/double-conversion source files into a new source/ subdirectory, to segregate upstream files from update.sh and our local patches. r=froydnj
--HG--
rename : mfbt/double-conversion/LICENSE => mfbt/double-conversion/source/LICENSE
rename : mfbt/double-conversion/README => mfbt/double-conversion/source/README
rename : mfbt/double-conversion/bignum-dtoa.cc => mfbt/double-conversion/source/bignum-dtoa.cc
rename : mfbt/double-conversion/bignum-dtoa.h => mfbt/double-conversion/source/bignum-dtoa.h
rename : mfbt/double-conversion/bignum.cc => mfbt/double-conversion/source/bignum.cc
rename : mfbt/double-conversion/bignum.h => mfbt/double-conversion/source/bignum.h
rename : mfbt/double-conversion/cached-powers.cc => mfbt/double-conversion/source/cached-powers.cc
rename : mfbt/double-conversion/cached-powers.h => mfbt/double-conversion/source/cached-powers.h
rename : mfbt/double-conversion/diy-fp.cc => mfbt/double-conversion/source/diy-fp.cc
rename : mfbt/double-conversion/diy-fp.h => mfbt/double-conversion/source/diy-fp.h
rename : mfbt/double-conversion/double-conversion.cc => mfbt/double-conversion/source/double-conversion.cc
rename : mfbt/double-conversion/double-conversion.h => mfbt/double-conversion/source/double-conversion.h
rename : mfbt/double-conversion/fast-dtoa.cc => mfbt/double-conversion/source/fast-dtoa.cc
rename : mfbt/double-conversion/fast-dtoa.h => mfbt/double-conversion/source/fast-dtoa.h
rename : mfbt/double-conversion/fixed-dtoa.cc => mfbt/double-conversion/source/fixed-dtoa.cc
rename : mfbt/double-conversion/fixed-dtoa.h => mfbt/double-conversion/source/fixed-dtoa.h
rename : mfbt/double-conversion/ieee.h => mfbt/double-conversion/source/ieee.h
rename : mfbt/double-conversion/strtod.cc => mfbt/double-conversion/source/strtod.cc
rename : mfbt/double-conversion/strtod.h => mfbt/double-conversion/source/strtod.h
rename : mfbt/double-conversion/utils.h => mfbt/double-conversion/source/utils.h
extra : rebase_source : e84171373b2f766698ea48d23484ed05463a4d8a
2017-01-23 15:11:50 -08:00
Jeff Walden 70018b9399 Bug 1332797 - Clear out the current double-conversion copy before updating it so that our copying properly handles added/removed files. r=froydnj
--HG--
extra : rebase_source : b4a4b088dfe3a1d08cc1600c04da5dc65e484807
2017-01-23 15:10:47 -08:00
Jeff Walden ad91d7c540 Bug 1332797 - Make the double-conversion update script clone the double-conversion repository into a temporary directory, then copy out of it, taking an optional git revision to use. After this patch, |./update.sh 04cae7a8d5ef3d62ceffb03cdc3d38f258457a52| is a no-op. r=froydnj
--HG--
extra : rebase_source : 6fab09db9cec6aaba6004014d32c93338127bd26
2017-01-23 11:27:01 -08:00
Steve Fink 7700214282 Bug 1264053 - Transfer DifferentProcess ArrayBuffers by copying, r=jorendorff
--HG--
extra : rebase_source : aed39bb2f92888af7626fd4c37df366cb1761bb8
extra : histedit_source : 4e231e7ef1b0b21d0c4bff2ebaa611e8b321e6d4
2017-01-19 14:02:40 -08:00
Botond Ballo 79e41b270d Bug 1331495 - Add an operator<< overload for Maybe. r=froydnj
MozReview-Commit-ID: 8TcYoikeCvR

--HG--
extra : rebase_source : 591e3c635ae5521418105e2ad1fc2672143eceeb
2017-01-18 13:53:35 -05:00
John Paul Adrian Glaubitz ba58645c87 Bug 1329194 - mfbt:tests: Define RETURN_INSTR for SH in TestPoisonArea. r=glandium 2017-01-20 09:12:22 +09:00
Tom Ritter 4d9bda0d08 Bug 1317642 Disable WeakPtr thread assertions on MinGW r=froydnj
MinGW has two threading models: win32 API based, which disables std::thread,
and POSIX based which enables it but requires an emulation library (winpthreads).

Rather than attempting to switch to pthread emulation at this point, we are
disabling the std::thread based assertion checking for WeakPtr on MinGW.

MozReview-Commit-ID: BmHo70n6AuK

--HG--
extra : rebase_source : 08495775b2925a797c8098216911d30c0b79ef3d
2017-01-13 13:50:21 -06:00
David Major e46dbc2c0f Bug 1324093 - Part 4: Make MOZ_Report{Crash,AssertionFailure} be MOZ_NEVER_INLINE. r=froydnj
You'd think that this would throw off the assertion stacks in nsTraceRefcnt::WalkTheStack. But as far as I can tell, it was already setting |skipFrames| too high!

On top of that, the function was getting out-of-lined in some instances already. It really should have been MOZ_ALWAYS_INLINE_EVEN_DEBUG.

MozReview-Commit-ID: J2FZmi0pKro

--HG--
extra : rebase_source : 20e5be9f8c21637a28435f47b8ab2de101825679
2017-01-18 09:37:54 +13:00
David Major 0cf4adb5f3 Bug 1324093 - Part 3.5: Add MOZ_MAYBE_UNUSED to mfbt/Attributes.h. r=froydnj
MozReview-Commit-ID: Ef1nkRipaDg

--HG--
extra : rebase_source : ea298dc948a027e5a686c2f022cce5d8cb170f27
2017-01-18 09:37:54 +13:00
David Major 2166409a1d Bug 1324093 - Part 3: Don't MOZ_ReportAssertionFailure in non-debug builds. r=froydnj f=Waldo f=nbp
Not only does this trim the code, it also makes MOZ_RELEASE_ASSERT follow the advice of MOZ_CRASH earlier in the file:

 * If we're a DEBUG build and we crash at a MOZ_CRASH which provides an
 * explanation-string, we print the string to stderr.  Otherwise, we don't
 * print anything; this is because we want MOZ_CRASH to be 100% safe in release
 * builds, and it's hard to print to stderr safely when memory might have been
 * corrupted.

MozReview-Commit-ID: Kuxzn1v9Vfs

--HG--
extra : rebase_source : 5c6efe7cb9adb1c366b423d6ff8f95002512985c
2017-01-18 09:37:46 +13:00
David Major 77329182eb Bug 1324093 - Part 2: Don't AnnotateMozCrashReason on debug builds. r=bsmedberg
I left gMozCrashReason visible (but not meaningfully used) in all builds, in order to match the behavior of Assertions.cpp, and to avoid more #ifdef clutter in nsExceptionHandler.cpp.

MozReview-Commit-ID: smoFkddGzd

--HG--
extra : rebase_source : 498f927f62fc944edf254c2ff3b115131367a506
2017-01-18 09:34:02 +13:00
David Major dd8c1f9d4e Bug 1324093 - Part 1: Move MOZ_REALLY_CRASH's null-deref and TerminateProcess into a never-inline function. r=froydnj
The C versus C++ distinction was only there so that Android could make sure it used the global ::abort. I didn't see the need to maintain the distinction for Windows. (Besides, with this change we're no longer doing textual inclusion of "TerminateProcess" in the macro, so people can't take over the name.)

Linux's abort sequence wasn't long enough to be troublesome, so I left it alone.

MozReview-Commit-ID: Ah5XtWpevGz

--HG--
extra : rebase_source : 37c3fb4c50bcba8e48c6a965a02e3f8608940538
2017-01-18 09:33:25 +13:00
David Major d27150b889 Bug 1324093 - Part 0: Use MOZ_{BEGIN,END}_EXTERN_C in Assertions.h. r=froydnj
Drive-by cleanup.

MozReview-Commit-ID: 8slQlbS3NS2

--HG--
extra : rebase_source : 2df96c531b8cfa03ca17c134393c5f054ca25927
2017-01-18 09:32:47 +13:00
James Cheng 642927f52b Bug 1329334 - Variant::as<> should use MOZ_RELEASE_ASSERT. r=froydnj
MozReview-Commit-ID: E95959RYm2r

--HG--
extra : rebase_source : 774fcf05c3e6baf080a0e74358e7441242465bb4
2017-01-10 14:24:49 +08:00
Nicholas Nethercote ceb22bead5 Bug 1325541 - Expand comments for MOZ_MUST_USE. r=froydnj.
DONTBUILD because it's a comment-only change.

--HG--
extra : rebase_source : 1b2343279f15a05ba633a0407f3eba07d6231083
2016-12-23 16:11:33 +11:00
David Major bbf01a2d5f Bug 1329236: Remove the DECLTYPE(expr) workaround in TestMaybe.cpp. r=ehsan
MozReview-Commit-ID: 5n5iRdJKn7M

--HG--
extra : rebase_source : 181165142b1f39c59c404b0b692383dd63a423b9
2017-01-06 12:53:20 -06:00
Nathan Froyd f6a73ed989 Bug 1315274 - rename mozilla::MakeRange to mozilla::IntegerRange; r=Waldo
MakeRange is just way too generic for this sort of thing.
2017-01-06 09:22:53 -05:00
Jason Orendorff 5d60d5ca87 Bug 1312001 - Scramble hash codes securely, to avoid leaking bits of object and symbol addresses.
MozReview-Commit-ID: yR1cIjrlPP

--HG--
extra : rebase_source : 871821e53eee5502cd255d52f02665f6845e3f09
2016-11-30 15:31:56 -06:00
Chris Peterson 20bc668a02 Bug 1280295 - Remove MOZ_HAVE_REF_QUALIFIERS. r=froydnj 2017-01-03 22:42:33 -08:00
Xidorn Quan b4ceaa3474 Bug 1325351 part 3 - Make Maybe accept value from different Maybe type when the inner type is convertible. r=froydnj
MozReview-Commit-ID: 2kYTncYh1Or

--HG--
extra : rebase_source : f3263ce10f3afae6e8ac734296dbb0e641c55c99
2016-12-23 11:49:33 +11:00
Xidorn Quan 30961dadd7 Bug 1325351 part 2 - Simplify 'Some' function. r=froydnj
MozReview-Commit-ID: AkLVQw9WYwy

--HG--
extra : rebase_source : 97449d0b2376d53ee7761da2a1ab425492c70ba6
2016-12-23 11:32:33 +11:00
Xidorn Quan 3526b1e58d Bug 1325351 part 1 - Remove workaround for VC in Maybe. r=froydnj
MozReview-Commit-ID: Anlrf3D2WtN

--HG--
extra : rebase_source : 9cc2e29e374753ad40774ebd0a0efce9f9449e3b
2016-12-29 10:58:01 +11:00
Sebastian Hengst 82a8828349 Backed out changeset b4d189ca8add (bug 1325351) for spidermonkey cgc and plain shell failure in Maybe.h. r=backout 2016-12-27 10:41:01 +01:00
Sebastian Hengst fe7c0739e3 Backed out changeset 79754b9b89b7 (bug 1325351) 2016-12-27 10:39:49 +01:00
Sebastian Hengst 6c5fadc6f5 Backed out changeset 45f0755ff471 (bug 1325351) 2016-12-27 10:39:45 +01:00